EFF: The stockholders of a bankrupt firm may get wiped out in a reorganization, but there is some probability that things will improve and the stock will eventually be in the money. The language is purposeful. The stock is like an out-of-the-money call option. The value of such an option is positive unless the probability of eventually being in-the-money is literally zero.

KRF: This is a great question because the limits of arbitrage are unusually important for these securities. As Gene says, in some cases there is a chance stockholders will receive a payment when the firm emerges from bankruptcy. Without frictions, I would expect the stock price to equal the present value of the expected payoff. Stocks are typically delisted when a firm files for bankruptcy, so their prices are not in the CRSP database most academics use to study stock returns. Probably as a result, I have not seen any studies of the in-bankruptcy performance of publicly traded stocks. There is a reasonable chance, however, that if we did look at their returns we would conclude the stocks of bankrupt firms underperform. In other words, I suspect that, on average, in-bankruptcy stock prices exceed the present value of their expected payoff.

How could the reported prices be too high on average? We usually think wise arbitrageurs will step in and short a security if the price is too high. The problem here is that an arbitrageur can tie up a lot of capital shorting in-bankruptcy securities. Recall that an arbitrageur must borrow the stock before he can short it and, when he does, he must post collateral to guarantee he will return the shares. If the stock price falls below $1, as is common in bankruptcy, the industry standard is to use a price of $1 when determining collateral. For example, if the stock price is $0.10 and the arbitrageur shorts 1 million shares, he receives only $100 thousand from the sale but he must post collateral worth perhaps $1.2 million.

Even more important, the arbitrageur is exposed to substantial buy-in risk. Security lenders can recall their shares at any time. In liquid markets, this is not a big concern because the borrower can cover his position by finding another lender or, at worst, buying the shares in the market. But after bankrupt firms are delisted, their equity is traded on the Pink Sheets and tends to be highly illiquid. As a result, if his loan is called, it may be difficult for the arbitrageur to find a new lender, and it will almost certainly be expensive to buy the shares in the market.

Because of these frictions, arbitrageurs will stop shorting an in-bankruptcy stock that is almost certainly worthless well before they drive the price to zero. Of course, if all investors were rational and used all available information about a firm and its trading environment when valuing equity, these limits of arbitrage would be irrelevant. But it is hard to believe that all investors share that level of rationality.

General Motors may illustrate the point. The SEC and FINRA issued an alert advising investors that the old GM stock, now trading under the ticker MTLQQ, will almost certainly be worthless when the company completes its liquidation. On its own website, the company warns, "Management continues to remind investors of its strong belief that there will be no value for the common stockholders in the bankruptcy liquidation process, even under the most optimistic of scenarios." Despite these warnings, as of this writing the stock is trading at $0.58 per share. Of course, there is some chance the SEC, FINRA, and management are all wrong and shareholders will eventually receive substantially more than $0.58 — but it seems unlikely.